Expression of Wnt signaling skeletal development genes in the cartilaginous fish, elephant shark (Callorhinchus milii)

Highlights

• Identification of genes that encode for β-catenin, Sfrp and SOST/SOSTDC1 in cartilaginous fish.

• The elephant shark β-catenin, Sfrp and SOST/SOSTDC1 genes are expressed in cartilage.

• Elephant shark β-catenin protein is localized in skeletal tissues.

Abstract

Jawed vertebrates (Gnasthostomes) are broadly separated into cartilaginous fishes (Chondricthyes) and bony vertebrates (Osteichthyes). Cartilaginous fishes are divided into chimaeras (e.g. ratfish, rabbit fish and elephant shark) and elasmobranchs (e.g. sharks, rays and skates). Both cartilaginous fish and bony vertebrates are believed to have a common armoured bony ancestor (Class Placodermi), however cartilaginous fish are believed to have lost bone. This study has identified and investigated genes involved in skeletal development in vertebrates, in the cartilaginous fish, elephant shark (Callorhinchus milii). Ctnnb1 (β-catenin), Sfrp (secreted frizzled protein) and a single Sost or Sostdc1 gene (sclerostin or sclerostin domain-containing protein 1) were identified in the elephant shark genome and found to be expressed in a number of tissues, including cartilage. β-catenin was also localized in several elephant shark tissues. The expression of these genes, which belong to the Wnt/β-catenin pathway, is required for normal bone formation in mammals. These findings in the cartilaginous skeleton of elephant shark support the hypothesis that the common ancestor of cartilaginous fishes and bony vertebrates had the potential for making bone.

Introduction

Our understanding of the human genome has benefited greatly from comparative studies with other vertebrate genomes (Aparicio et al., 1997, Gilligan et al., 2002, Venkatesh et al., 2006). Gene sequences that give rise to particular phenotypes can be identified by comparison with closely related genomes, while comparisons with distantly related genomes are valuable for identifying evolutionarily conserved regulatory elements that play essential roles in development and physiology (Amores et al., 1998, Aparicio et al., 1997, Gilligan et al., 2002, Venkatesh et al., 2006, Yu et al., 2008).

The skeletal structure within vertebrates is made of different tissues that are dynamic and multifunctional (Hall, 2005). Bone, cartilage, dentine and enamel together are structural, supportive, protective, aid articulation, act as storage sites and help maintain mineral homeostasis (Bilezikian et al., 2008, Hall, 2005). The developmental pathways of each of these tissues have evolved to serve the specific needs of individual species to allow them to flourish within their environmental niches. Aquatic animals have a constant supply of calcium and other ions, while land-dwelling vertebrates have to obtain their calcium through dietary consumption (Danks and Richardson, 2011). A bony skeleton has been considered a feature of highly evolved vertebrates including the most successful extant class, Mammalia. Bone not only provides physical support but also acts as a readily accessible reservoir of calcium (Singh, 1998). However, many vertebrates including jawless fish, sharks and rays have a predominantly cartilaginous skeleton with only regions of calcified tissue known as endoskeletal tesserae (Kemp and Westrin, 1979). The endoskeletal calcification is different to that of endochondral ossification and remains tesserate throughout life which is one of the main features distinguishing these cartilaginous fish (Chondrichtyes) from teleosts and higher vertebrates (Kemp and Westrin, 1979). The tesserate pattern of discontinuous calcified regions separated by uncalcified cartilage provides advantages of structural support together with flexibility compared to that offered by a bony skeleton (Kemp and Westrin, 1979).

Cartilage was considered the most primitive vertebrate skeletal tissue, however this idea was reversed when paleontological evidence proved bone was already present in the skeletons of the earliest vertebrates, Ostracoderms (Berrill, 1955). Bone is regarded as the more adult skeletal tissue, and that ancestors of modern sharks could form endoskeletal bone but during evolution lost their ability to make bone and a more “degenerate” form, cartilage, exists in sharks today (Moss, 1977, Romer, 1942). Romer (1942) maintained that all fish, including sharks, had a common jawed vertebrate ancestor with bony armour (Class Placodermi). The remnants of this armour, form the scales on modern fish, and the dermal denticles on the skin of many species of sharks (Hall, 2005, Kemp and Westrin, 1979). However, cartilage does not readily fossilize and only calcified cartilage has been identified in one of the most primitive heterostracan agnathans (extinct class of jawless vertebrates) known from the fossil record (Orvig, 1951). Claims of bone being more primitive than cartilage, disregarding the possibility of these two skeletal tissues evolving simultaneously and the link between chondrogenesis to endochondral ossification (calcification of cartilage to bone) (Goldring et al., 2006) remains unjustified.

The analyses of fish genome sequences have provided useful information for understanding the structure, function and evolution of vertebrate genes and genomes (Venkatesh, 2003, Venkatesh et al., 2007, Yu et al., 2008). Teleost fish (e.g. zebrafish and pufferfish) and cartilaginous fish (e.g. sharks, skates, rays and chimaeras) are among the most distantly related vertebrates to humans. Direct comparisons of the human and teleost genomes are complicated due to a ‘fish-specific’ whole genome duplication event that occurred in the ancestors of the teleost fish, resulting in many duplicated gene loci in teleosts (Amores et al., 1998, Christoffels et al., 2004, Prohaska and Stadler, 2004). For example, mammals contain four Hox gene clusters (HoxA, HoxB, HoxC, and HoxD) containing 39 Hox genes (Kuraku and Meyer, 2009, Venkatesh et al., 2007). Hox genes are important for development as they express specific transcription factors that control anterior–posterior tissue patterning along the body axis of animals (Kuraku and Meyer, 2009). Following genome duplication teleost fish contained two copies of each Hox gene cluster. However, the extent of this gene duplication varies in different teleost lineages and the rates at which specific duplicated genes have mutated vary significantly among different teleost fish lineages (Amores et al., 1998, Aparicio et al., 1997, Prohaska and Stadler, 2004). Pufferfish and zebrafish contain 7 Hox gene clusters containing 45 and 49 Hox genes respectively (Amores et al., 1998, Aparicio et al., 1997, Christoffels et al., 2004, Prohaska and Stadler, 2004). However, pufferfish have lost one copy of the HoxC cluster, while zebrafish have lost one copy of the duplicated HoxD cluster. Thus, it is not always straightforward to determine relationships between the genes of teleosts and humans.

Cartilaginous fish represent the oldest living phylogenetic group of jawed vertebrates (Benton and Donoghue, 2007). Consequently, they are an important group for our understanding of the developmental and physiological systems of jawed vertebrates, and for identifying specialized genomic features that have contributed to the divergent evolution of higher vertebrates (Fig. 1) (Venkatesh et al., 2007, Yu et al., 2008). Cartilaginous fish provide a critical reference in attempting to reconstruct the evolutionary history of vertebrate genomes (Venkatesh et al., 2007).

The elephant shark (Callorhinchus milii) has been used as a vertebrate reference genome, following the generation of a 1.4× shotgun genome sequence (∼75% coverage) by Venkatesh et al. (2007). In contrast to other cartilagionous fish (elasmobranchs) which contain only 3 Hox clusters (A, B, D) (King et al., 2011), elephant shark contain 4 Hox clusters (A, B, C, D) similar to mammals (Amores et al., 1998, Aparicio et al., 1997, Prohaska and Stadler, 2004, Venkatesh et al., 2007). The level of conserved non-coding elements (CNEs) has also been compared between genomes (Venkatesh et al., 2006). CNEs are generally associated with transcription factor genes and developmental genes, and are enriched for enhancers directing expression to specific domains during embryonic development. The elephant shark has twice as many CNEs conserved with the human genome as teleosts (Venkatesh et al., 2006). Furthermore, the elephant shark genome has a higher level of conserved synteny with the human genome compared to teleost-human genome comparisons, as teleost fish genomes have a higher level of rearrangements resulting from the whole-genome duplication they experienced (Venkatesh et al., 2007, Yu et al., 2008). The highly conserved non-coding regions in the elephant shark genome compared with the divergent regulatory regions in teleosts underscore the importance of the elephant shark as a critical reference vertebrate genome (Venkatesh et al., 2006, Venkatesh et al., 2007, Venkatesh et al., 2005, Yu et al., 2008). Survey sequencing of the elephant shark genome has provided useful information regarding the small size (910 Mb), gene complement and organisation of its genome (Venkatesh et al., 2007). It has also highlighted specific examples of vertebrate genes and gene families that have been lost differentially in mammalian and teleost lineages (Venkatesh et al., 2007). We are therefore using the elephant shark as a model cartilaginous fish to investigate the origin of genes involved in the formation of a bony/cartilaginous skeleton.

Differential patterns of expression and localization of the proteins involved in skeletal tissue formation have enabled the evolution of the skeleton between different species allowing them to occupy different environmental niches. The Wnt/β-catenin pathway was identified as having an important role in skeletal development when naturally occurring mutations within the pathway were identified as causing disorders of bone mass in vertebrates (Babij et al., 2003, Blish et al., 2008, Boyden et al., 2002, Hartmann, 2007, Hartmann and Tabin, 2000, Issack et al., 2008, Westendorf et al., 2004). This pathway is highly conserved throughout evolution but its role in skeletal development is not fully understood (Hartmann, 2007, Issack et al., 2008, Moon et al., 2002). Wnt signaling not only promotes bone formation but also plays a role in maintaining cartilage (Day et al., 2005, Hartmann and Tabin, 2000). Differential expression of the Wnt/β-catenin pathway genes plays an important role in cell differentiation to either bone or cartilage (Day et al., 2005, Hartmann, 2007, Hartmann and Tabin, 2000). Wnt signaling has been demonstrated to either negatively or positively regulate osteoblast and chondrocyte differentiation (Bodine, 2008, Chen and Alman, 2009, Chun et al., 2008, Day et al., 2005, Hartmann, 2007, Hartmann and Tabin, 2000). A β-catenin-mediated Wnt signal accelerates the maturation process of chondrocytes to endochondral ossification during bone formation (Day et al., 2005, Zhu et al., 2009), while a β-catenin independent Wnt-signaling pathway delays chondrogenesis (cartilage development) (Chun et al., 2008).

There are several antagonists to the Wnt/β-catenin pathway. One of the major extracellular groups of antagonists are the secreted frizzled related proteins (SFRPs) which contain the ligand binding domain of the Wnt co-receptor Frizzled (Fzd) but lack the transmembrane region (Bafico et al., 1999, Kawano and Kypta, 2003). These proteins can bind Wnt but cannot transduce the signal and thus effectively block Wnt signaling. Sclerostin is an important negative regulator of Wnt signaling that exclusively inhibits bone formation (Chan et al., 2011, Krause et al., 2010, Semenov et al., 2005). In mature mammals, sclerostin is expressed in osteocytes, and suppresses bone formation by inhibiting the differentiation of osteoblasts (Krause et al., 2010, Poole et al., 2005, van Bezooijen et al., 2004, Winkler et al., 2003). Sclerostin inhibits Wnt/β-catenin signaling by binding to the transmembrane low density lipoprotein receptor-related protein (LRP) (Semenov et al., 2005, van Bezooijen et al., 2004, van Bezooijen et al., 2005). Sclerostin is also thought to function as a bone morphogenetic protein (BMP) antagonist (van Bezooijen et al., 2005, Winkler et al., 2003).

In this study, we hypothesized that the elephant shark expresses the genes and corresponding proteins comprising key elements of the Wnt/β-catenin pathway. We have identified elephant shark sequences highly similar to those of the human genes of the Wnt/β-catenin pathway encoding β-catenin (Ctnnb1), secreted frizzled related protein (Sfrp) and sclerostin (Sost). We also determined the tissue-specific mRNA expression and protein localization of these members of the Wnt/β-catenin pathway and show that these patterns are similar to those occurring in human tissues. Comparison of the genes and corresponding proteins of the Wnt/β-catenin pathway between other vertebrates and the cartilaginous elephant shark could provide further insight into the evolution and functioning of this pathway, elucidating how Wnt signaling can control both bone and cartilage skeletal development. Analysis of the regulation of expression of genes in this pathway in the elephant shark provides a simpler system for studying components of the complex Wnt signaling pathways and their role in skeletal development.